Silks are fibrous protein secretions that exhibit exceptional strength and toughness and as such have been the target of extensive study. Silks are produced by over 30,000 species of spiders and by many insects. Very few of these silks have been characterised, with most research concentrating on the cocoon silk of the domesticated silkworm, Bombyx mori and on the dragline silk of the orb-weaving spider Nephila clavipes. 
In the Lepidoptera and spider, the fibroin silk genes code for proteins that are generally large with prominent hydrophilic terminal domains at either end spanning an extensive region of alternating hydrophobic and hydrophilic blocks (Bini et al., 2004). Generally these proteins comprise different combinations of crystalline arrays of β-pleated sheets loosely associated with β-sheets, β-spirals, α-helices and amorphous regions (see Craig and Riekel, 2002 for review).
As silk fibres represent some of the strongest natural fibres known, they have been subject to extensive research in attempts to reproduce their synthesis. However, a recurrent problem with expression of Lepidopteran and spider fibroin genes has been low expression rates in various recombinant expression systems due to the combination of the repeating nucleotide motifs in the silk gene that lead to deleterious recombination events, the large gene size and the small number of codons used for each amino acid in the gene which leads to depletion of tRNA pools in the host cells. Recombinant expression leads to difficulties during translation such as translational pauses as a result of codon preferences and codon demands and extensive recombination rates leading to truncation of the genes. Shorter, less repetitive sequences would avoid many of the problems associated with silk gene expression to date.
In contrast to the extensive knowledge that has accumulated about the Lepidopteran (in particular the cocoon silk of Bombyx mori) and spider (in particular the dragline silk of Nephila clavipes) little is known about the chemical composition and molecular organisation of other insect silks.
In the early 1960s, the silk of the aculeate Hymenopteran was shown to have an alpha-helical structure by X-ray diffraction patterns obtained from silk fibres drawn from the salivary gland of honeybee larvae (Rudall, 1962). As well as demonstrating that this silk was helical, the patterns obtained were indicative of a coiled-coil system of alpha-helical chains (Atkins, 1967). Similar X-ray diffraction patterns have been obtained for cocoon silks from other Aculeata species including the wasp Pseudopompilus humbolti (Rudall, 1962) and the bumblebee, Bombus lucorum (Lucas and Rudall, 1967).
In contrast to the alpha-helical structure described in the Aculeata silks, the silks characterised from a related clade to the aculeata, the Ichneumonoidea, have parallel-β structures. X-ray diagrams for four examples of this structure have been described in the Braconidae (Cotesia(=Apenteles) glomerate; Cotesia(=Apenteles) gonopterygis; Apenteles bignelli) and three in Ichneumonidae (Dusona sp.; Phytodietris sp.; Bronchus femoralis) (Lucas and Rudall, 1967). In addition the sequence of a single Braconidae (Cotesia glomerate) silk has been described (Genbank database accession number AB188680; Yamada et al., 2004). This partial protein sequence consists of a highly conserved 28 X-asparagine repeat (where X is alanine or serine) and is not predicted to contain coiled coil forming heptad repeats. Extensive analysis of the amino acid composition of the cocoon silks of the Braconidae has shown that the silks from the subfamily Microgastrinae are unique in their high asparagine and serine content (Lucas et al., 1960; Quicke et al., 2004). Related subfamilies produce silks with significantly different amino acid compositions suggesting that the Microgastrinae silks have evolved specifically in this subfamily (Yamada et al., 2004). The partial cDNA of Cotesia glomerata was isolated using PCR primers designed from sequence obtained from internal peptides derived from isolated cocoon silk proteins. The predicted amino acid composition of this partial sequence closely resembles the amino acid composition of the extensively washed silk from this species.
The structure of many of the silks within other non aculeate Apocrita and within the rest of the Hymenoptera (Symphata) are most commonly parallel-β sheets, with both collagen-like and polyglycine silks produced by the Tenthredinidae (Lucas and Rudall, 1967).
Honeybee silk proteins are synthesised in the middle of the final instar and can be imaged as a mix of depolymerised silk proteins (Silva-Zacarin et al., 2003). As the instar progresses, water is removed from the gland and dehydration results in the polymerisation of the silk protein to form well-organised and insoluble silk filaments labelled tactoids (Silva-Zacarin et al., 2003). Progressive dehydration leads to further reorganisation of the tactoids (Silva-Zacarin et al., 2003) and possibly new inter-filamentary bonding between filaments (Rudall, 1962). Electron microscope images of fibrils isolated from the honeybee silk gland show structures of approximately 20-25 angstroms diameter (Flower and Kenchington, 1967). This value is consistent with three-, four-, or five-stranded coiled coils.
The amino acid composition of the silks of various aculeate Hymenopteran species was determined by Lucas and Rudall (1967) and found to contain high contents of alanine, serine, the acid residues, aspartic acid and glutamic acids, and reduced amounts of glycine in comparison to classical fibroins. It was considered that the helical content of the aculeate Hymenoptera silk was a consequence of a reduced glycine content and increased content of acidic residues (Rudall and Kenchington, 1971).
Little is known about the larval silk of the lacewings (Order: Neuroptera). The cocoon is comprised of two layers, an inner solid layer and an outer fibrous layer. Previously the cocoon was described as being comprised of a cuticulin silk (Rudall and Kenchington, 1971), a description that only related to the inner solid layer. LaMunyon (1988) described a substance excreted from the malphigian tubules that made up the outer fibres. After deposition of this layer, the solid inner wall was constructed from secretions from the epithelial cells in the highly villous lumen (LaMunyon, 1988).
It is also known that lacewing larva produce a proteinaceous adhesive substance from the malpighian tubules throughout all instars to stick the larvae to substrates, to glue items of camouflage on to the larvae's back or to entrap prey (Speilger, 1962). In the genus Lomamyia (Bethothidae), the larvae produce the silk and adhesive substance at the same time and it has been postulated that these two substances may well be the same product (Speilger, 1962). The adhesive secretion is highly soluble and is also thought to be associated with defense against predators (LaMunyon & Adams, 1987).
Considering the unique properties of silks produced by insects such as Hymenopterans and Neuropterans, there is a need for the identification of novel nucleic acids encoding silk proteins from these organisms.